the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Jul 2026
Dynamical Mechanism behind the Southern Hemisphere Westerly Intensification during the Last Glacial Maximum: A Linkage between Sea-ice and Polar Front Jet
Abstract. The Southern Hemisphere westerlies (SHW) play a pivotal role in modulating the global carbon cycle and climate feedback. However, their behavior during the Last Glacial Maximum (LGM) is debated owing to discrepancies between paleoclimate models and proxy records. While tropical upper-tropospheric cooling and Antarctic surface cooling exert opposing influences on the SHW, the detailed dynamical mechanisms through which Antarctic sea-ice expansion modulates large-scale atmospheric circulation are poorly understood. In this study, we investigated the dynamical mechanisms of austral winter SHW change under altered orbital and surface conditions with a series of climate model simulations. By conducting sensitivity experiments with varying Antarctic sea-ice concentrations, we isolated the thermodynamic effect of sea ice from the tropical cooling signal. Our results demonstrate that sea-ice-induced surface cooling drives the poleward intensification of the SHW through two distinct mechanisms. First, strong surface cooling steepens the meridional temperature gradient near the sea-ice edge, thereby directly maintaining the SHW intensity through thermal wind balance. Second, the enhanced baroclinicity amplifies eddy heat fluxes and storm track activity. The resulting increase in storm track activity drives a downward transfer of upper-tropospheric westerly momentum, reinforcing the surface westerlies. Through these mechanisms, the sea-ice-driven cooling outweighed the opposing equatorward influence of tropical cooling. This study provides a dynamical framework for understanding how sea-ice thermodynamic forcing drives large-scale circulation changes in the context of LGM climate conditions.
- Preprint
(4469 KB) - Metadata XML
-
Supplement
(785 KB) - BibTeX
- EndNote
Status: open (until 02 Sep 2026)
- RC1: 'Comment on egusphere-2026-3834', Anonymous Referee #1, 22 Jul 2026 reply
-
RC2: 'Comment on egusphere-2026-3834', Anonymous Referee #2, 25 Jul 2026
reply
Title: Dynamical Mechanism behind the Southern Hemisphere Westerly Intensification during the Last Glacial Maximum: A Linkage between Sea-ice and Polar Front Jet
Author(s): Hyeong-Gyu Kim, Joowan Kim, Sang-Yoon Jun, Seong-Joong Kim, and Damwon So
MS No.: egusphere-2026-3834Summary
This paper investigates why the Southern Hemisphere westerly winds (SHWs) strengthen and shift poleward during the Last Glacial Maximum (LGM), focusing on the role of expanded Antarctic sea ice. Using a series of climate-model sensitivity experiments, the authors seek to separate the effects of Antarctic sea-ice expansion from the competing influence of tropical cooling. Overall, the study reinforces the view that Antarctic high-latitude cooling can contribute to a poleward intensification of the SHWs, potentially outweighing a possible tendency for tropical cooling to produce an equatorward response.
The key contribution is not demonstrating that sea ice affects the westerlies—that has been suggested previously—but providing a more detailed dynamical framework linking Antarctic sea-ice expansion, surface cooling, enhanced polar-front temperature gradients, increased storm-track activity, and stronger, more poleward Southern Hemisphere westerlies.
Major comments
Whilst this is a useful and potentially valuable analysis of the mechanisms linking Antarctic sea-ice change to SHWs, I have three major concerns that need to be addressed. Given these are rather major, it seems not useful at this stage to focus on the smaller presentation and writing issues - these are omitted from this review.
1. Lack of engagement with palaeoclimate constraints and observations
My most significant concern is that the study is insufficiently grounded in palaeoclimate evidence. There is very little evaluation against LGM sea-ice and SST reconstructions, and no attempt to compare the simulated circulation changes with available wind-sensitive or moisture-sensitive proxy records.
This omission is important because the long-standing debate over LGM Southern Hemisphere westerlies arises specifically from differences between model behaviour and palaeoclimate interpretations. Without a more detailed discussion of observational constraints, it is difficult to assess whether the simulated wind responses are realistic. At present, the study remains largely detached from the palaeoclimate problem that motivates it.
I therefore strongly encourage the authors to place their simulations within the context of available LGM sea-ice, SST, moisture, and atmospheric circulation reconstructions, and to discuss whether the imposed surface changes and simulated response are consistent with existing proxy evidence.
2. Problems with the imposed surface forcing experiments
The sensitivity experiments are useful for isolating mechanisms, but I am not convinced that the current experimental design is sufficient to support the broader conclusions being drawn regarding the role of Antarctic sea ice in driving LGM wind changes.
The principal issue is that the expanded sea-ice experiments impose substantial sea-ice changes on unchanged piControl SSTs. Whilst such experiments are useful as idealised perturbations, they are not physically representative of LGM surface conditions, where expanded Antarctic sea ice would have been accompanied by substantial local and regional Southern Ocean cooling.
To support the central interpretation of the manuscript, an additional set of experiments should test whether the diagnosed response remains robust when SST changes are dynamically consistent with the imposed sea-ice expansion. In particular, experiments with combined Antarctic sea-ice and polar/subpolar SST changes appear necessary.
A related issue is that the manuscript repeatedly discusses the role of tropical cooling, yet this mechanism is largely inferred rather than directly tested. If the authors wish to conclude that Antarctic cooling dominates over tropical cooling in determining the SHW response, then targeted tropical SST sensitivity experiments are required. At present, this comparison remains incomplete
3. Understanding and referencing of the existing literature
The manuscript cites several relevant studies; however, it does not sufficiently engage with earlier work that addresses many of the same questions.
In particular, the authors should read - and then discuss at several points in their manuscript:
Sime, L.C., Kohfeld, K.E., Le Quéré, C., Wolff, E.W., de Boer, A.M., Graham, R.M., and Bopp, L. (2013). Southern Hemisphere westerly wind changes during the Last Glacial Maximum: model–data comparison. Quaternary Science Reviews, 64, 104–120.
This study addresses several key aspects that are either absent or only partially treated in the present manuscript. First, it places simulated Southern Hemisphere westerly wind changes within the context of LGM palaeoclimate observations, including a systematic comparison against a large synthesis of SST, sea ice, moisture proxy data and discussion of other relevant palaeo-environmental constraints. Second, it explicitly investigates the sensitivity of Southern Hemisphere westerlies to individual LGM boundary conditions, including Antarctic sea ice, Southern Ocean SSTs, tropical sea surface changes, greenhouse gases, orbital forcing, and ice-sheet changes. The study identified sea-surface temperature/ice changes, particularly those associated with enhanced Southern Ocean temperature gradients, as a dominant control on simulated wind changes, while also demonstrating an important role for tropical cooling.
As a result, Sime et al. (2013) addresses many of the broader questions that motivate the present manuscript. I therefore encourage the authors to position their contribution more clearly relative to this earlier work, particularly with regard to observational constraints and the interpretation of sensitivity experiments.
4. Final comments on framing
In my view, the principal advance of the present study is not the identification of Antarctic sea ice as a control on LGM westerlies, nor the distinction between tropical and high-latitude forcing – which the current experiments do not do. Rather, the major contribution is the more detailed dynamical diagnosis linking sea-ice expansion, enhanced baroclinicity, storm-track activity, eddy momentum transport, and strengthening of the polar-front jet. Framing the manuscript more explicitly in this way – alongside considering model-data comparison exercises in any final discussion - would help clarify its novelty and significance.
Citation: https://doi.org/10.5194/egusphere-2026-3834-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 96 | 27 | 10 | 133 | 26 | 10 | 11 |
- HTML: 96
- PDF: 27
- XML: 10
- Total: 133
- Supplement: 26
- BibTeX: 10
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General Comments
This paper uses the CESM1.2 Earth System model to examine the mechanisms controlling shifts in the position and intensity of the Southern Hemisphere Westerly winds during the Last Glacial Maximum (LGM). It first examines changes in the SHW between the LGM and pre-industrial (PI) simulations and then conducts a series of sensitivity tests to examine the effects of step-wise sea-ice expansion on changes in the SHW.
The paper demonstrates that the CESM 1.2 model simulates a poleward shift and intensification of the SHW during the LGM, via two mechanisms: (a) the intensification of the meridional temperature gradient, and (b) enhanced baroclinity that intensifies eddy heat fluxes and storm tracks. The authors argue that these factors outweigh the competing factor driving equatorward shifts in the SHW (i.e. cooling of the upper-troposphere in the tropics).
My concern is with the set-up of the sensitivity experiments, which superimpose sea-ice changes on an atmosphere-only PI simulation where the SSTs are prescribed to match modern ocean conditions. The authors use these conditions so that they are solely examining the effects of sea-ice expansion; however, we know that sea-surface temperatures during the LGM averaged 1.5-5˚C cooler in the Southern Ocean and were as much as 6-8˚C in parts of the subantarctic zone (see Figure 3, Kohfeld et al., 2013). These substantially cooler SSTs would undoubtedly reduce the surface meridional temperature gradients called upon in these experiments to drive a poleward intensification of the SHW through amplified atmospheric baroclinicity and increased eddy activity. As such, I think that the design of the sensitivity experiments artificially enhances a surface meridional temperature gradient to invoke the anticipated result.
RECOMMENDATION 1: I suspect that the authors could get around this problem with one additional sensitivity experiment in which they select one of their sensitivity tests (e.g., ICE75) and run it again with LGM SSTs prescribed, at least in the Southern Hemisphere. This would allow the authors to isolate the effects of the artificial surface meridional T gradient imposed by their choice of prescribed conditions.
RECOMMENDATION 2: Aside from this issue, the analysis of the LGM-PI simulations is interesting and seems sound to me. That said, I suspect that the authors may come up against some challenges, because there is a strong, very vocal contingent in the paleo-community who strongly believes that the SHW shifted equatorward during the LGM. I believe that the data in Kohfeld et al. (2013) are open enough that a poleward shift is possible. The authors have been careful to focus on mechanisms, but I think their argument would be strengthened if they conducted some form of comparison with paleodata, as was done in Sime et al. (2013), showing that even a poleward shift in the westerlies could be consistent with the moisture proxies originally used to infer an equatorward shift in the SHW.
Specific Comments
METHODS: Based on Sime et al. (2016) it seems important to know how the CESM 1.2 simulates the modern sea-ice extent. Can you provide a one-sentence clarification?
METHODS: It would be helpful to state in the text the latitudinal expansion of the sea-ice extent in the sensitivity experiments – these values look like they are imbedded in Figure 1 but it would be good to have a one-sentence addition stating them explicitly in the text.
FIGURE CAPTIONS: It would be useful to indicate in the Figure Captions which figures are using the fully couples earth-system model and which are based on the sensitivity experiments.
LN 422-424: Many readers/subsequent authors overlook the fact that Kohfeld et al. (2013) left the interpretation of the paleo-data quite open. They actually concluded that a 3-5˚ equatorward expansion was ONE POSSIBLE interpretation of the paleo data. Kim et al. make this point earlier in the paper (and should here as well); there are several OTHER interpretations that are consistent with the paleodata reconstructions.
Technical Corrections
Ln 20: Do you mean surface or upper troposphere cooling?
STRUCTURE: It is unusual and strange to have a section entitled “Summary and Conclusions” BEFORE you begin the Discussion section. I recommend either (a) retitling the section “Summary” and making it part of Section 3, or (b) expanding the section and moving it AFTER the Discussion.
LN 444-446: I’m not actually sure which model-proxy discrepancy you are referring to here. Can you be more specific, because you haven’t actually done any data-model comparisons in this paper.